Research On Key Technologies Of Complex System Simulation

Simulation of complex systems is currently one of the most important methods to study complex systems. However, due to the complex relations within the system complexity, uncertainty, the overall behavior of the emergence of other reasons make us different from the usual simple simulation system research. Complex systems need to consider the level of simulation, heterogeneous, widely distributed geographic characteristics of subsystems, comprehensive simulation system design. Therefore, the characteristics of this article will demand complex system simulation starting, simulation of complex systems for key technologies in-depth study and build versatility and scalability to support complex system simulation environment. The main research contents are as follows:Firstly, from the characteristics of complex systems simulation, analyzes the distributed simulation system architectural requirements, based on the proposed layered system architecture, hardware aspects of the composition of the system, simulation applications, software environment, human-computer interaction, The simulation system is divided into: the hardware layer, data layer, business logic, presentation layer, etc., and highlights the business logic layer simulation support platform specific features.On the basis of the simulation system level architecture design study on the integration of multi-granularity model, simulate scheduling, and fault injection were respectively studied. For the multi-level granularity standardized management model proposed model specification and model integration division size specification, in two specifications, based on the model proposed collection and integrated approach, and the establishment of a mechanism for the indexing model calls.Traditionally, task allocation and task scheduling designed separately, may cause simulation application needs and available resources mismatch. For dependencies simulation tasks, sort tasks, assign node is a prerequisite for task scheduling. Based on the task sorting and distribution algorithm, for centralized scheduling strategy to make improvements, and proposed scheduling algorithm and node synchronization algorithms.Fault injection as an effective response test system behavior and reliability of technology it has been widely used in the engineering system. Therefore, on the basis of distributed simulation system architecture based on the proposed model of fault-based injection means, and the corresponding time-driven fault simulation algorithm. Finally, an example to verify the feasibility and effectiveness of the interface of the method.Finally, to build distributed simulation support system and the establishment of specific examples of simulation applications, through simulation data and system performance test data analysis to verify the accuracy of the simulation system, versatility, scalability and simulation of complex systems feasibility and effectiveness of key technologies.